Like Texas, With AI "Everything is Bigger"

In many ways, vendor financing of artificial intelligence infrastructure is a bit like Texas: “everything’s bigger.”

Nobody knows yet whether “circular financing” is going to be a major problem in the artificial intelligence business, but it’s reaching new levels. 

Nvidia, for example, is pondering commitments to OpenAI of about $600 billion, including:

• An OpenAI Ohio data center lease financial guarantee of $250 billion 

• Separately, financing another $350 billion of GPU purchases for OpenAI. 

If completed, that would represent one of the largest examples of vendor-supported infrastructure finance in technology history.

Vendor financing has been provided by companies such as Cisco, Lucent, IBM, and GE Capital in the past, but not at such scale.

But Nvidia has increasingly used several mechanisms to support customers beyond simply shipping chips.

Customer	Approximate size	Nvidia role	Similarity to Ohio deal
OpenAI (Ohio campus)	Project >$500B; reported $250B guarantee plus possible $350B GPU financing	Credit guarantee, GPU financing, hardware supplier	Most extensive
OpenAI (2025 infrastructure agreement)	Up to $100B investment commitment	Infrastructure investment tied to deployment of Nvidia systems	High (Fierce Network)
CoreWeave	Multi-billion-dollar	Equity investor; guaranteed purchases of unused cloud capacity	High (Reuters)
CoreWeave	Multiple equity rounds	Early strategic investor before IPO	Medium (Reuters)
xAI	Tens of billions in GPU systems	Large hardware supplier; strategic ecosystem partner	Moderate (Reuters)
Oracle / Stargate	Hundreds of billions of AI infrastructure	Hardware supplier and infrastructure partner	Moderate (SSRN)
Numerous AI startups	Hundreds of millions to billions	Venture investments through NVentures plus preferred GPU access	Lower, but follows same ecosystem strategy (NVIDIA)

To some extent, Nvidia’s moves are an example of how various contestants in the AI value chain are staking claims in broader roles within the value chain. High-performance computing services suppliers such as Amazon and Google create their own chips and sponsor or create their own language models.

So it might not be surprising to see Nvidia taking on new roles as well. 

Function	Nvidia role
GPU supplier	Sell chips
Systems supplier	Sell complete AI clusters
Platform company	CUDA, networking, software, AI factories
Capital provider	Equity investments, financing, guarantees, demand commitments

The reported Ohio arrangement is not simply a very large chip sale. 

It would make Nvidia part supplier, part infrastructure financier, and part credit guarantor.Nvidia has previously invested in customers such as CoreWeave and OpenAI,  and has used demand guarantees and equity investments to accelerate AI infrastructure.

But such financing has been a staple of the computing industry since the time of mainframes. 

Vendor financing has been a recurring feature of the computing industry for more than 60 years. It tends to emerge during periods when a new generation of computing requires exceptionally large up-front investment. 

The mechanism changes over time, from leases to loans to equity investments to purchase guarantees.

But the economic logic remains consistent: If customers cannot afford the infrastructure needed to create the next wave of demand, suppliers help finance that infrastructure.

The reported Nvidia/OpenAI proposal is best understood as the latest version of this long-running pattern.

Era	Dominant technology	Financing mechanism	Strategic purpose
1960s–1970s	Mainframes	Leasing	Reduce customer capital burden
1980s	Minicomputers	Vendor credit	Expand installed base
1990s	Enterprise networking	Vendor financing	Accelerate Internet buildout
2000s	Telecom & hosting	Vendor loans, export finance	Support infrastructure expansion
2010s	Cloud computing	Long-term purchase commitments	Enable hyperscale investment
2020s	AI infrastructure	Equity, guarantees, GPU financing	Accelerate AI ecosystem growth

AI infrastructure is so capital-intensive that financing has returned to center stage.

Supplier	Customer	Financing approach	Circular element
Nvidia	CoreWeave	Equity investment plus demand guarantees	Nvidia helps create GPU demand
Nvidia	OpenAI	Reported credit guarantees and GPU financing	Financing supports purchases of Nvidia GPUs
AMD	Various AI cloud providers	Strategic investments and joint development (smaller scale)	Encourages accelerator adoption
Microsoft	OpenAI	Multi-billion-dollar investments tied to Azure usage	Investment generates Azure revenue
Amazon	Anthropic	Multi-billion-dollar investment tied to AWS usage	Investment drives AWS consumption
Google	Anthropic	Large investment tied to Google Cloud	Investment increases cloud demand

History suggests such financing can work. But history also suggests it can fail. We still do not know what the AI outcome will be. 

Condition	IBM	Cisco	Nvidia
Technology creates lasting productivity gains	✔	✔	Likely
Customers eventually generate sustainable cash flow	✔	Mixed	Unknown
Vendor does not assume excessive credit risk	✔	No	Still uncertain

Across six decades, the industry has repeatedly followed the same sequence:

• A breakthrough technology emerges (mainframes, PCs, the Internet, cloud, AI)

• Infrastructure costs initially exceed customers' ability or willingness to pay

• Suppliers devise financing mechanisms to accelerate adoption

• If demand proves durable, the financing is remembered as visionary

• If demand disappoints, the same financing is criticized as excessive risk-taking.

The reported Nvidia–OpenAI arrangement is unprecedented in scale, but not in principle. The novelty lies less in the existence of vendor financing than in its magnitude: guarantees and financing measured in the hundreds of billions of dollars rather than millions or even billions.

For Every Public Purpose There is a Corresponding Private Interest

For every public purpose there are corresponding private interests. OpenAI, for example, has its own interests in language models and the data center infrastructure necessary to support widespread use of such models.

Where it comes to public policy on data centers, that applies to state actors as well.

A report issued by OpenAI discusses two clusters of ChatGPT accounts likely originating from China

“that we banned after they used our models in support of apparent covert influence operations that promoted narratives in an attempt to manipulate a legitimate debate about American AI and wider tech policies,” the report says. 

“The first cluster generated social media comments and images claiming that data center buildouts for AI were increasing electricity prices for average families,” the report says. The issue is not the relevance of the debate, but the effort by state-linked actors to influence that debate, for perceived state interests.

“The operators of the accounts were likely part of a social media operations team at a private Chinese technology company conducting work for Chinese provincial level government clients,” OpenAI says. “This activity appears consistent with a commercial ecosystem that supports Party-state priorities in public opinion guidance.”

“A separate report they uploaded to ChatGPT described their objectives and strategies for influencing public opinion and establishing social media accounts designed to evade platform detection systems,” the report notes.

“They primarily targeted U.S. audiences and generated English-language short comments and images claiming that data centers and AI applications were increasing electricity demand and causing higher

costs for ordinary Americans,” OpenAI says. 

While the campaign does not appear to have gained much authentic engagement, “their significance lies in what they reveal about the intentions of influence operators from China and the narratives they are testing and seeking to amplify,” the report states.

First Amendment Free Speech Protections Apply Only to Government, at the Federal Level, Not Private Firms

Some might believe that U.S. free speech rights apply to most venues we encounter, from shopping malls to public transportation. That is mistaken. The right of free speech as enumerated in the First Amendment to the U.S. constitution only fully applies to government restraint, not that of all private actors or venues. 

So when a passenger on an airline complains about free speech “rights,” that is not a venue of protected speech.

Importantly, the First Amendment generally restricts government action, not private property owners, so it usually does not give a right to speak on private property or in privately run venues just because the public is invited in.

The Supreme Court’s starting point is that private owners do not have to turn their homes, businesses, or other property into expressive forums for others. In Marsh v. Alabama, the Court treated a company town like a municipality because it functioned as a town in all practical respects, so speech rights applied there. 

But later cases narrowed that idea, making clear that ordinary private commercial property is usually not subject to First Amendment speech access rights.

Consider shopping malls, which some litigants have compared to the older “town square.” Courts have ruled that the First Amendment does not force the owner to allow leafleting, protests, or petitioning on private mall property, as in Lloyd Corp. v. Tanner and Hudgens v. NLRB.

But some state constitutions grant greater speech rights in shopping centers than the federal First Amendment requires.

Social media companies are usually treated as private actors, so the First Amendment generally does not make them open public forums for user speech, either. 

Public transportation is trickier because the answer depends on ownership and operation. If the transit system is run by the government, First Amendment limits on viewpoint discrimination can apply because the government is involved. 

If the service or property is privately owned or operated, the First Amendment usually does not itself force access, although specific transit areas can sometimes be treated as public forums for certain government-run advertising or station spaces.

Churches are private property and are not generally subject to First Amendment speech-access claims from outsiders. 

The general rule is that the First Amendment applies directly when the government is restricting speech, but not when a private owner is doing so. 

Language Model Brand Preference is Unclear, Yet

Whether language model usage represents sustained brand preference is probably unclear. One study found satisfaction rankings for the top three platforms (Claude, ChatGPT, and DeepSeek) statistically indistinguishable. 

That study also found that users treat these tools as interchangeable utilities rather than sticky ecosystems. More than 80 percent of respondents use two or more platforms, and switching costs are negligible. 

Also, each platform attracts users for different reasons: ChatGPT for its interface, Claude for answer quality, DeepSeek through word-of-mouth, and Grok for its content policy, the researchers say. 

So usage patterns, though clear enough, should not necessarily equate to sustainable brand preference, for the moment. 

In consumer markets, ChatGPT continues to hold the lead, but Gemini has climbed significantly. 

source: Momentic Marketing 

In the enterprise user market, Anthropic’s Claude leads, followed by OpenAI and Gemini. 

source: Momentic Marketing 

Still, most enterprises aren’t betting on a single model provider, as some 81 percent now use three or more model families in testing or production, according to Andreessen Horowitz. 

Model leadership also varies by use cases. Anthropic is notable for enterprise software development use cases, for example. ChatGPT gets wide use for general-purpose chatbot use cases. 

a16z.com

Trust, compliance, and "authority" positioning matter more to business buyers than to consumers. To the extent we can say brand positioning exists, Anthropic seems to be the brand for enterprises seeking “reliability.” 

In consumer markets, perhaps Gemini's growth is tied heavily to being pre-installed rather than actively chosen by users, and OpenAI's own remaining enterprise strength is partly inertia.

But the basis for brand preference seems clear enough in enterprise markets. Only about 11 percent of enterprise teams reported switching vendors in the past year. Most (66 percent) upgraded the same vendor's model, so Anthropic's enterprise gains represent real vendor displacement, not just organic growth of new spend.

At least at this point, though, language model brand preference durability seems less pronounced than was the case for earlier computing innovations:

• Operating systems and browsers historically showed durable lock-in from network effects, file-format compatibility, and default-setting. Once Windows or Chrome won a segment, share moved in single-digit points per year for a decade.

• Enterprise infrastructure (databases, cloud, ERP) traditionally locks in through multi-year contracts, data gravity, and integration cost, producing switching costs high enough that even a technically superior challenger struggles for years. LLM enterprise switching costs are comparatively low. 

• Multi-homing is common, as  low switching costs and task-specific model preferences seem to exist.

So language model brand preference today functions less like durable ecosystem lock-in (OS, browsers, smartphones) and more like a fluid, feature- and task-driven preference market, except perhaps in enterprises, where trust, compliance positioning, and demonstrated coding performance have already produced one real reversal of market leadership in under three years.